Jump to content

Majority function

From Wikipedia, the free encyclopedia
This is an old revision of this page, as edited by David Eppstein (talk | contribs) at 21:16, 24 November 2009 (explicitly credit Valiant with the monotone formula for majority and give more explicit bounds on its size). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

In Boolean logic, the majority function (also called the median operator) is a function from n inputs to one output. The value of the operation is false when n/2 or more arguments are false, and true otherwise. Alternatively, representing true values as 1 and false values as 0, we may use the formula

The "−1/2" in the formula serves to break ties in favor of zeros when n is even; a similar formula can be used for a function that breaks ties in favor of ones.

A major result in circuit complexity asserts that this function cannot be computed by AC0 circuits of subexponential size.

A majority gate is a logical gate used in circuit complexity and other applications of Boolean circuits. A majority gate returns true if and only if at least 50% +1 of its inputs are true.

Monotone formulae for majority

For n = 1 the median operator is just the unary identity operation x. For n = 3 the ternary median operator can be expressed using conjunction and disjunction as xy + yz + zx. Remarkably this expression denotes the same operation independently of whether the symbol + is interpreted as inclusive or or exclusive or.

For an arbitrary n there exists a monotone formula for majority of size O(n5.3) (Valiant 1984).

Properties

Among the properties of the ternary median operator < x,y,z > are:

  1. < x,y,y > = y
  2. < x,y,z > = < z,x,y >
  3. < x,y,z > = < x,z,y >
  4. < < x,w,y > ,w,z > = < x,w, < y,w,z > >

An abstract system satisfying these as axioms is a median algebra.

References

  • Valiant, L. (1984), "Short monotone formulae for the majority function", Journal of Algorithms, 5: 363–366, doi:10.1016/0196-6774(84)90016-6.
  • Knuth, Donald E. (2008). Introduction to combinatorial algorithms and Boolean functions. The Art of Computer Programming. Vol. 4.0. Upper Saddle River, NJ: Addison-Wesley. pp. 64–74. ISBN 0-321-53496-4.

See also